THE ANALYSIS OF COMPLEX LEARNING ENVIRONMENTS (New chapter for second edition of Rethinking pedagogy for a digital age.

) Peter Goodyear & Lucila Carvalho University of Sydney, Australia

INTRODUCTION: ANALYSIS FOR DESIGN ! Pedagogy, as the art and science of helping other people learn, can be practiced in a variety of ways, including through direct face-to-face teaching. Our work seeks to understand and inform pedagogy that is enacted more indirectly as design for learning that is, where people committed to facilitating other peoples learning carry out their work primarily through the design of worthwhile learning tasks and/or the design of appropriately supportive learning resources. Given the focus of this book, our attention is on designs in which digital resources play a significant part, though we will argue that design often works best when it takes a more holistic approach working with networks of interacting digital and non-digital entities. In this chapter, our focus is on analysis for design. Analysis connects with design in a number of ways, e.g. through needs analysis a classic starting point for a structured design process (Crandall et al. 2006). It also plays a major role in evaluation informing judgements at the end of a design cycle, about whether something is working well, and about what might need to be improved (Reigeluth and Carr-Chellman, 2009).

Our approach to analysis is rather different. Its distinctiveness arises from these four observations. (1) Design rarely takes place on a green field site. Analysis needs to be able to capture what exists already, not just what success would look like. Design activity can then make proposals that will work within, and improve upon, an existing

set of constraints and possibilities. (2) Many things affect any single learning process. So analysis must be able to represent a complex array of influences, some of which are human, some physical (including digital). (3) Competence, which is one way of describing the end goal for a learning process, rarely resides in the head of a learner. Rather, a persons competence is usually entangled in, and dependent on, a set of social and physical relationships such that a more expansive view of competence includes that persons ability to assemble and hold together the entities needed for the task at hand. When analysis is used to create a description of competence, or of a desired state of affairs a smoothly working system - it needs to be able to deal with such complexity. (4) Since a number of influential models of learning involve some kinds of apprenticeship, authentic engagement in practice, legitimate peripheral participation, experiential learning, etc., then the kind of description created by analysis in (3) is needed if designers are to see what else they may need to help set in place to support such processes of learning through engagement in practice.

We employ the term learning environment with some trepidation, since it is widely used but rarely explained in writing about learning technologies. Its a term that appears to work neatly when the focus is on an individualised learner and their physical environment, side-stepping questions about whether it is reasonable to describe other people as part of ones environment, or whether learning environment can be used to describe the (shared) habitat of a collection of learners (Goodyear, 2000a). As will become clear, our use of the term is relational person and environment are mutually entailed; there is no person without an environment and no environment without a person (or organism) dwelling in it (Ingold, 2000).

"!

SHIFTING THE FOCUS OF ANALYSIS FROM DISCRETE DEVICES TO ECOLOGIES AND NETWORKS Later in this chapter we sketch two contrasting learning situations to illustrate our argument for a more systemic approach to analysis. Both situations involve a number of digital technologies, as well as other elements that combine to have educationally consequential effects. Some of these elements were designed for the situation, some were selected by teachers or designers, some had other ways of coming to hand when students were engaged in their work. A thorough analysis needs ways of identifying and connecting diverse kinds of elements, including the physical, digital and human; texts, tools and artefacts; tasks, rules and divisions of labour. Producing an inventory of components is not enough, because functioning depends on structure on relationships between parts. Nor does it make much sense to try to identify the contribution to learning made by a single component. Outcomes depend on interactions between multiple entities. Rather, we need forms of analysis that match the complexity of contemporary learning challenges holistic, ecological or architectural rather than fragmenting, reductionist modes of thought.

We argue that richer, socio-material analyses of learning environments provide knowledge that fits well with the needs of design and designers. We suggest that providing better ways of thinking about analysis, evaluation and design can help dislodge unhelpful habits of thought especially those that try to isolate intrinsic merits of particular tools, media or pedagogies. We also argue that such analysis sharpens perception of the boundary between what can be designed, and what must emerge at learntime.

#!

Our approach to analysis shifts the focus from individual elements of an educational innovation to the system level (e.g. Luckin 2010; Boys, 2011; Ellis and Goodyear 2010; Westberry and Franken, in press). It starts by recognising that learning activity takes place in complex, messy, dynamic situations, in which interactions between elements produce conditions that are more or less supportive of learning. knowledge generation [is] a joint exercise of relational strategies within networks that are spread across space and time, and performed through inanimate (e.g. books, mobile phones, measuring instruments, projection screens, boxes, locks) as well as animate beings in precarious arrangements Learning and knowing are performed in the processes of assembling and maintaining these networks, as well as in the negotiations that occur at various nodes comprising a network Things not just humans, but the parts that make up humans and nonhumans persuade, coerce, seduce, resist and compromise each other as they come together. (Fenwick et al. 2011, p10) Actor Network Theory (ANT), as used by Fenwick and colleagues, is one of a number of perspectives that can be used to try to capture some of this complexity. We remain agnostic about some of the key ideas associated with ANT such as whether it is reasonable to attribute agency to artefacts. But, like other schools of thought implicated in the materialist turn, ANT sensitises us to the ways in which material objects influence human activity (see also Boivin, 2008; Sorensen, 2009; Bennett, 2010; Fenwick and Edwards, 2010; Miller, 2010; Johri 2011). It reminds us that matter matters. Ecological psychology similarly challenges presumptions about the superiority of mind over matter (Gibson 1977, 1986). From Gibsons work, educational technology has appropriated one of its core and most contested concepts the idea that objects have affordances which shape the behaviour of people who encounter them (Laurillard 1987; Norman 1999; Conole and Dyke 2004; Oliver 2005, 2011; Turner 2005; John and Sutherland 2005; Dohn 2009). Affordance does a great

$!

deal of work in educational technology partly because it sidesteps issues about technological determinism without suggesting that technology choices can be arbitrary. But as Harry Collins has observed: the terms afford and affordance are lazy terms these terms merely paper over deep cracks in our understanding of why, given the extraordinary interpretive capabilities of humans, anything affords any one interpretation better than any other something hidden and mysterious is going on whenever the terms afford and affordance make their appearance (Collins 2010, p36) We will come back to this issue shortly. For now, the key points are as follows: (1) analysing or evaluating learning activity in context cannot sensibly be reduced to enumerating the pedagogical affordances of individual tools, devices or artefacts; (2) instead, a more systemic approach is needed, in which learning and the things that influence it are seen as connected in webs or assemblages; (3) how we conceptualise the functioning of the web has serious consequences for how we analyse and explain what happens affordance turns out to be just one of several necessary terms.

DESIGN AND ITS PRODUCTS ! Much of the learning that students do is accomplished without direct supervision. In such circumstances, with only very limited opportunities for teachers to carry out realtime repairs, good design is crucial. Since analysis and design need a shared conceptual framework, if they are to be mutually informing, then we offer the following sketch of design and its legitimate products. It consists of three broad principles, each of which is unpacked in a subsequent section (see Goodyear, 2000b; Goodyear and Retalis 2010, for further information).

%!

1. Design for learning is chiefly concerned with the design of good learning tasks (well-crafted suggestions of good things for people to do, if they are to achieve some desired learning outcome). 2. Design for learning must also attend to the social and physical setting ensuring (as far as is possible) that all the resources needed for learning come to hand. 3. Design for learning needs to work fluently across scale levels: linking macro, meso and micro. ! 1) Task design typically results in the production of texts often in the form of a specification of what should be done. Students interpret these texts and their subsequent learning activity can be understood as an improvisation that is informed (but rarely determined) by the text. It is often through their interpretation of key texts (such as course outlines and assignment specifications) that students unravel what is required from them in a given situation. This is rarely a straightforward process. On the one hand students will bring their own beliefs and experiences about how such activity is to be completed (Biggs & Tang, 2007; Prosser & Trigwell, 1999; Ellis & Goodyear, 2010) and their ability to keep task specifications in mind, as activity unfolds, will be constrained by working memory. On the other hand it is necessary that students are able to recognise and realize the relevant meanings associated with the pedagogic/learning context they are in (Bernstein 2000). Students interpretation of what should be done including the designers intentions - requires that they are able to identify implicit social values associated with knowledge and practices within a particular context. That means, students will need to translate a number of cues that are communicated to them along with the text. (Such cues can appear in a variety of modes including verbal, written, images or through implicit and subtle signs in the learning environment.) Once students understand the rules of the game, they then may be able to act and produce (texts and other artefacts) according to what is

&!

expected from them. Some of these implicit social values reflect underlying organizing principles structuring knowledge in particular fields of practice (Maton 2000; Carvalho, Dong and Maton 2009). They underlie the way in which pedagogical communication takes place, regulating teachers/designers practices and shaping for example, the ways a task is proposed. Consequently, task design also involves incorporating ways of expressing the broader social context of the proposed learning activities so that students know the rules for the context they are in. In short, tasks as designed objects - need to be understood as (a) nested in an architecture of tasks (tasks make sense in relation to sub-tasks and supra-tasks), and (b) located within what might be called an epistemic architecture a structure of knowledge and ways of knowing peculiar to a discipline, profession or practice. ! 2) Design that is attending to the physical and social setting(s) within which learning activity is expected to unfold typically results in the identification, selection, recommendation and/or creation of texts, tools and artefacts that the designer believes will be useful. It also results in suggestions to students about how they might work with others proposing divisions of labour, grouping, and/or the allocation of roles. As with task specifications, these socio-material design components should normally be understood as resources on which students may choose to draw even when their use is mandated, students find themselves some wriggle room (Goodyear and Ellis 2010). Moreover, working with and in a complicated network of people, tools, artefacts and places is neither an automatic nor a dependable process: what works needs to be seen as an accomplishment (Law & Mol, 2002; Rabardel & Beguin, 2005).

'!

3) Design for learning gravitates towards the meso-level (Jones et al. 2006). By this we mean that, in practice, educational design attention tends to be drawn to the design of learning tasks that run over hours or days, rather than years or seconds. It is better aligned to the layout of rooms or the recommendation of specific texts than to macro considerations (replanning the campus; restocking the library) or to the minutiae of students choices of pen, paper, or workmate. That said, the devil can often be in the detail and also macro-level phenomena can place powerful constraints on what happens at the meso-level. So while design tends to focus on the meso, it cannot safely ignore chains of influence that run from macro to micro and back again. The inter-relations between tools, artefacts and other material/digital resources for learning can be thought of as constituting a physical architecture. Similarly, inter-personal working relationships, divisions of labour, roles etc make sense within what might be called a social architecture. ! In sum, whether we are trying to analyse an existing learning situation, or design a new one, we need ways of conceiving of the networks of interacting people, objects, activities, texts etc that shape learning activities and outcomes. We need to be able to detect global forces at work in local artefacts, and to account for the mutual shaping done by language, minds and things (Gibson & Ingold, 1995; Miller, 2010).

CONNECTING ASSEMBLAGES OF TOOLS AND ARTEFACTS TO HUMAN ACTIVITY ! An analysis of the relations between such things as digital tools and resources (on the one hand) and learning outcomes (on the other) needs to be informed by some defensible ideas about how the former can be said to influence the latter. Causality is multiple and complex. Evaluative or analytic approaches that assume simple linear ! (!

causal paths are unlikely to be helpful. How then to frame analysis of learning environments so that we stand some chance of connecting (a) that which is designed to (b) valued educational outcomes? If one finds it sufficient to equate learning with authentic engagement in a social practice, then this is a one step argument. If one also values some associated change in the understanding or skills of a learner, then two steps are needed to complete the connections (x-ref Learning chapter).

We take an activity-centered position on this: what matters is what the learner does physically, mentally and emotionally (Shuell 1992; Biggs and Tang 2007). Different kinds of knowledge are acquired in different ways through the activation of different kinds of mental processes for example (Ohlsson 1995, 2011). So the nature of the learners activity is part of the link between the material world and their learning outcomes. The other missing link is between the material world and activity. This is where the over-used idea of affordance is normally asked to weave its magic. It is usually a mistake to try to isolate some intrinsic properties of tools, resources, places etc and connect them to learning. Rather, as Nicole Boivin argues:

material properties are always properties relative to people, as James Gibsons concept of affordances reminds uswhat is important is not just materiality, but the coming together of materiality and embodied humans engaged in particular activities. (Boivin 2008, p167, emphasis added).

The quotation we took from Harry Collins (above), about affordance being a lazy term, was arguing that the extraordinary interpretive capabilities of people undermine the explanatory power of affordance. It is true that people are extremely versatile sense-makers, but that does not mean that they linger in interpretive mode prior to

)!

every action. What needs to be acknowledged here is that human action can involve deliberation and interpretation but it can also be rapid, fluid and seemingly automatic. Rather than insist on the primacy of either affordance or interpretation in explaining relations between material objects and human activity, we would argue that both play a role, much of the time. One way to think about this is consider Daniel Kahnemans argument that humans rely on two systems of mental operation tuned to thinking fast and slow. Kahneman (2011) describes two systems in the mind.

System 1 operates automatically and quickly, with little or no effort and no sense of voluntary control System 2 allocates attention to the effortful mental activities that demand it The operations of System 2 are often associated with the subjective experience of agency, choice and concentration (Kahneman 2011)

We suggest that affordances are involved when System 1 is running the show; interpretation invokes System 2. This immediately provides a more flexible and robust way of accounting for links between the material/digital world, learner activities and learning outcomes. For example, providing learners with scaffolding for their activities, by offering them guidance in the form of digital texts, necessarily invokes System 2, increases cognitive load, but also opens opportunities for interpretation and reflection on the designers intentions. Design can substitute other forms of computer-based guidance for texts e.g. through the use of interface icons, or other forms of procedural support, that afford one action rather than another. This allows System 1 to do what is needed, reducing cognitive load but sacrificing opportunities for reflection in order to expedite action. (Neither of these approaches is intrinsically better. Design involves trade-offs.)

*+!

This leads to a view of analysis that can hypothesise a variety of connections between the material/digital world, learner activity and outcomes involving various mixtures of affordance and interpretation; fast thinking and slow; visceral, behavioural and reflective responses, or hot and cool cognition (Norman, 2005; Thagard 2008). It also helps resolve thorny problems about technological determinism and human agency (Oliver 2011) since few, if any, encounters with technology are single-stranded.

Much of the literature that aims at explaining relations between technology and human action takes a social, cultural or semiotic view, within which characteristics of tools and artefacts are of little interest. In subsuming material studies into general semiotic and social paradigms, we highlight certain aspects of material meaning, but at the same time we occlude recognition of what makes material things different from words and signs indeed what makes material things really interesting in their own right. (Boivin 2008, p155, emphasis added) Such examples allow us more clearly to see how the actual physical properties of things rather than just the ideas we hold about them instigate change, by placing constraints on some activities and behaviours, and making possible, encouraging, or demanding, other types of behaviour. (op cit, p166, emphasis added) Like Boivin, we think that analysis needs to account for ways in which technology (and the material world more generally) influence human perception and action, without recourse to deterministic arguments. Objects in the material world carry physical properties such as their size, weight, shape, colour and temperature which may or may not have been intended as part of their design. Digital tools and artefacts affect a narrower range of senses, but have qualities which can change in an instant. We also need to acknowledge that embedded into the particular way any material

**!

object is designed is an intention of how form and function were to meet. The object itself thereby carries values from, and choices made in, the design process. Either way, through their physical properties and embodied intentions, designed objects will have an effect on human perception and action.

ILLUSTRATIONS: ANALYSING THE ARCHITECTURE OF PRODUCTIVE LEARNING NETWORKS

Illustration 1: Field-training of para-medics (iPads in the wild)

! This case study came to our attention when one of our part-time Masters students began discussing her ideas about a dissertation project. (We have changed a few details, to preserve anonymity.) Her original suggestion was that she might try to evaluate the effects on learning of the introduction of iPads the context being one of the courses in her School of Health Sciences. She sketched how she might do this with some students having School-provided iPads and others not. The first opportunity to do this would be on a field trip an exercise in which students who are learning to be paramedics would take part in the search for, and treatment and evacuation of, some people injured while hiking in the mountains. (This exercise has been run annually for a number of years. The casualties are played by actors. Qualified mountain rescue personnel take a major role in running the exercise. The iPads were new.) We did not encourage our Masters student to run the experiment that she had in mind. Rather, we suggested that, at least at first, she should take a more exploratory approach roving around while the exercise unfolded, making field notes, and trying to identify and describe as carefully as possible the mix of things that seemed to influence the activities and their outcomes.

*"!

Her field notes mentioned that the iPads were used (but not often). She also noted the use of: compasses, maps, GPS devices, torches, whistles, ropes, binoculars and walkie-talkies. These were just the tools for navigation and communication. Then there were stretchers, bandages, scissors, watches (for measuring a pulse), stethoscopes, medications and syringes objects involved in the initial treatment of the actor/casualties once they had been located. A reasonably complete account would also take in these actor/casualties (semi-skilled), the mountain rescue volunteers (very skilled), the tutors (semi-skilled) and the students (often lost, cold and confused). Obvious though it might seem, the design and evolution of this exercise also necessitated being in the mountains. The difficulties of traversing rough terrain, locating a casualty when hidden in a valley or by vegetation, coping with poor visibility and communicating without mobile phones all played a substantial role in the exercise. Proper clothing is also important. A conventionally-minded instructional designer might be forgiven for thinking that good boots and a warm, waterproof coat are things for the students to provide. But those who forgot these important items were unable to complete the exercise. And which instructional design guideline tells you that fingerless gloves are useful when trying to use an iPad on a cold mountain?

Among many other things, this example illustrates how important it is that students are able to recognise the relevant meanings of the pedagogical context in which they find themselves: relating ways of knowing with material circumstances. Pedagogical interactions on the mountain involved very different rules of the game compared to those in a normal classroom. Students had to be able to identify a different language and those who could not recognize the essential rules within this context were then unable to participate fully in the experience.

*#!

From the perspective of the organisers as the designers of the exercise, emphasis was placed on the technical knowledge associated with understanding drugs, first aid etc; on the use of technical devices (e.g. iPads), and on life saving procedures. These were seen as the essential knowledge for completing the exercise, and some issues related to the effects of the environment were overlooked, or their influence underestimated. The organisers assumed that key aspects of the knowledge needed to work effectively in the mountain environment would come from the students prior personal experience - and therefore, they should already know that boots and coats were essential elements, given the material circumstances. As a result, in spite of whatever knowledge they had about life-saving procedures, using technology remotely and so on, those who did not know about the need for coats, gloves and boots in rough terrain failed to learn much from the experience. ! ! ! ! !"#$%&'()*+,&%(-(./0,'(1%&%!( ( ( ( ( We cannot easily portray the whole network of tools, artefacts, activities, people and attributes of the physical terrain in a single image. But Figure 1 begins to capture some of the relationships involved in (more or less) successful execution of this field exercise.

*$!

Our analysis suggests that successful participation in the exercise involves: learning to use each tool, at least with sufficient fluency to be able to act according to the established protocols, but ideally with a level of automaticity that binds tool and action in a smooth flow integrating the use of the tools into a web of activity, involving smooth effective action, coordination with others, focus on the priority goals, etc. turning the individual and aggregate experiences of the exercise into learning that lasts. The point of the exercise, for each student, is not just to master the individual tools but to participate in the construction of a co!ordinated web of activity that can result in a successful rescue, minimising danger to participants, and leaving traces (in some kinds of memory) that mean doing something like this again will not feel entirely new. !

Illustration 2: Online learning for educational leadership

Our second case involves an online professional development program for school teachers taking on curriculum leadership positions. (Again, unimportant details have been altered to protect anonymity.)

A number of elements of the program are quite conventional. There is an online induction module, introducing the participants to the technology being used, to a number of key ideas about educational leadership, and to the overall scope and goals of the program. Through direct experience of the resources, teaching methods, user interface, tasks and collaborative learning activities that will be used in the main part of the program, participants have an opportunity to work out whether the program will

*%!

suit them, and whether they will be able to cope with its demands. 30% of participants quit during or immediately after the induction module.

The remaining participants then tackle 12 structured learning modules (SLMs), each of which introduces them to a set of ideas that the course team believes to be relevant to understanding curriculum leadership. Once the participants have completed four of these modules, they are allowed to join an online Community of Practice, within which they are encouraged to discuss issues with peers. Once all the SLMs are completed, participants work in small groups to design curriculum implementation projects that they will carry out in their own schools. The designs are peer-reviewed. The rubric for the peer review includes criteria that reflect and encourage the use of concepts, techniques etc that were presented in the SLMs. Thus far, some 200 projects have been designed and published for peer review by the program participants.

An analysis of what is working well and what might be improved would conventionally focus on the quality of the resources being made available in the SLMs, the timeliness and helpfulness of online tutors support, the ease of use of the online tools, participants experiences and their assessment of the extent and usefulness of their own learning. All of these are important, but they tell less than the whole story, and an analysis of the case that was restricted to these elements would not (we contend) provide an adequate basis for others to design similar educational programs.

Not least, the fact that this program draws on problems that emerge in participants own educational practice and is intended to help solve those problems means that

*&!

the participants schools (in all their complexity) have to be counted in as learning resources. People without access to such resources could not participate successfully in the program. Moreover, these school-based resources are outside the sphere of things that the program providers can design. (The program providers/designers can specify requirements e.g. that participants must have a leadership role with respect to curriculum change in some part of their schools work. But they cannot design these important parts of the network of activities, texts and resources on which participants will draw.)

Participants bring their school-based resources to the mix and associated with each of these resources is a specific set of underlying principles structuring knowledge practices. That is, knowledge practices within each school reflect implicit values within that specific context, which shape participants practices and the way they see leadership and curriculum. As various participants come into the pedagogical context of the online environment to exchange ideas about leadership and curriculum, they bring also their own beliefs and values, which will need to be negotiated with the beliefs and values of other participants, whose practices are shaped by their own experiences of their school-based resource. A participant with a background in Science may see knowledge practices in a different way than a participant with a background in Arts. Or a participants views about leadership may reflect their experiences at different levels of hierarchy, such as being a Coordinator or a Principal. The context of the experience may also be influenced by the complexities that shape working in a city school versus in a remote area, an established versus a new school and so on. As these participants come together to exchange notions about leadership and discuss curriculum issues within the online environment, they do

*'!

so from their own perspective, from where they are positioned within the field. The design of the pedagogical context where they interact needs to address these differences, acknowledging that diverse underlying values are likely to be present.

This use of the local working context as a resource for online learning is not uncommon in design for professional development, but we have found very little in the instructional design literature that helps capture or think about key issues here, other than in general terms.

CONCLUDING POINTS ! In this chapter, we have suggested that approaches to analysing complex learning environments will be more productive, and will align better with the knowledge needs of designers, if they help map the webs of heterogeneous elements that shape learning activity. In particular, we have argued that neither affordance nor interpretation provides a sufficient explanation for the connections between that which is designed and the learners activity. Our illustrations show how tasks (and activities) sit within nested architectures, such that what a person is doing at any one point only makes sense in relation to a web of other tasks (and activities), the accomplishment of which may well be distributed quite widely in time and space, and across the material, human and digital. We have also tried to show something of the complexity of the networks of tools, artefacts, places, practices, ways of knowing and inter-personal relationships that are implicated in designed learning situations. Successful designs for learning find ways of embracing this complexity. Sharp analytic skills help us understand such designs, and learn from them. !

*(!

ACKNOWLEDGEMENTS We gratefully acknowledge the financial support of the Australian Research Council through grant FL100100203: Learning, technology and design: architectures for productive networked learning. We also thank Helen Beetham for insightful comments on an earlier draft of this chapter.

*)!

REFERENCES Bennett, J. (2010) Vibrant matter: a political ecology of things, Durham: Duke University Press. Bernstein, B. (2000) Pedagogy, symbolic control, and identity: theory, research, critique (Revised ed.), Oxford: Rowan & Littlefield. Biggs, J. and Tang, C. (2007) Teaching for quality learning at university: what the student does (3rd ed.) Buckingham: Open University Press. Boivin, N. (2008) Material cultures, material minds: the impact of things on human thought, society and evolution, Cambridge: Cambridge University Press. Boys, J. (2011) Towards creative learning spaces: re-thinking the architecture of post-compulsory education, New York: Routledge. Carvalho, L., Dong, A. and Maton, K. (2009) Legitimating design: a sociology of knowledge account of the field Design Studies, 30(5): 483-502 Collins, H. (2010) Tacit and explicit knowledge. Chicago: The University of Chicago Press. Conole, G. and Dyke, M. (2004) What are the affordances of information and communication technologies? ALT-J: Research in Learning Technology, 12(2): 113-124. Crandall, B., Klein, G., & Hoffman, R. (2006). Working minds: a practitioner's guide to cognitive task analysis. Cambridge MA: MIT Press. Dohn, N. (2009) Affordances revisited: articulating a Merleau-Pontian view, International Journal of Computer-Supported Collaborative Learning, 4(2): 151170. Ellis, R. and Goodyear, P. (2010) Students' experiences of e-learning in higher education: the ecology of sustainable innovation, New York: Routledge Falmer. Fenwick, T., and Edwards, R. (2010) Actor network theory in education, London: Routledge. Fenwick, T., Edwards, R., and Sawchuk, P. (2011) Emerging approaches to educational research: tracing the sociomaterial, Abingdon: Routledge. Gibson, J. (1977) The theory of affordances, in R. Shaw & J. Bransford (Eds.), Perceiving, acting, and knowing: Toward an ecological psychology (pp. 67-82). Hillsdale, New Jersey: Lawrence Erlbaum Associates. Gibson, J. (1986) The ecological approach to visual perception, Hillsdale NJ: Lawrence Erlbaum Associates. Gibson, K. and Ingold, T. (eds.) (1995) Tools, language and cognition in human evolution, Cambridge: Cambridge University Press. Goodyear, P. (2000a). Environments for lifelong learning: ergonomics, architecture and educational design. In J. M. Spector & T. Anderson (Eds.), Integrated and Holistic Perspectives on Learning, Instruction & Technology: Understanding Complexity (pp. 1-18). Dordrecht: Kluwer Academic Publishers. ! "+!

Goodyear, P. (2000b) Seeing learning as work: implications for understanding and improving analysis and design Journal of Courseware Engineering, 2: 3-11. Goodyear, P. and Ellis, R. (2010) Expanding conceptions of study, context and educational design, in R. Sharpe, H. Beetham & S. de Freitas (Eds.) Rethinking learning for a digital age: how learners are shaping their own experiences (pp. 100-113), New York: Routledge. Goodyear, P. and Retalis, S. (2010) Learning, technology and design, in P. Goodyear & S. Retalis (eds.) Technology-enhanced learning: design patterns and pattern languages (pp. 1-28), Rotterdam: Sense Publishers. Ingold, T. (2000) The perception of the environment: essays in livelihood, dwelling and skill, Abingdon: Routledge. John, P. and Sutherland, R. (2005) Affordance, opportunity and the pedagogical implications of ICT, Educational Review, 57(4): 405-413. Johri, A. (2011) The socio-materiality of learning practices and implications for the field of learning technology, Research in Learning Technology, 19(3): 207-217. Jones, C., Dirckinck-Homfeld, L. and Lindstrom, B. (2006) A relational, indirect, meso-level approach to CSCL design in the next decade, International Journal of Computer-Supported Collaborative Learning, 2(2-3): 35-56. Kahneman, D. (2011) Thinking, fast and slow, New York: Farrar, Straus and Giroux. Laurillard, D. (1987) The different forms of learning in psychology and education, in J. Richardson, M. Eysenck & D. Warren Piper (eds.) Student learning: research in education and cognitive psychology (pp. 198-207) Buckingham: Open University Press. Law, J., & Mol, A. (Eds.). (2002). Complexities: social studies of knowledge practices. Durham: Duke University Press. Luckin, R. (2010) Re-designing learning contexts: technology-rich, learner-centred ecologies, New York: Routledge. Maton, K. (2000) Languages of legitimation: The structuring significance for intellectual fields of strategic knowledge claims, British Journal of Sociology of Education 21(2): 147-67. Miller, D. (2010). Stuff. Cambridge: Polity Press. Norman, D. (1999) Affordance, conventions and design, Interactions, 6(3), 38-43. Norman, D. (2005). Emotional design: why we love (or hate) everyday things. New York: Basic Books. Ohlsson, S. (1995) Learning to do and learning to understand: a lesson and a challenge for cognitive modelling, in P. Reimann & H. Spada (eds.) Learning in humans and machines: towards an interdisciplinary learning science (pp. 37-62), London: Pergamon. Ohlsson, S. (2011) Deep learning: how the mind overrides experience, Cambridge: Cambridge University Press. Oliver, M. (2005) The problem with affordance, E-Learning, 2(4). Oliver, M. (2011). Technological determinism in educational technology research:

"*!

some alternative ways of thinking about the relationship between learning and technology. Journal of Computer Assisted Learning, 27, 373384. Prosser, M., & Trigwell, K. (1999). Understanding learning and teaching: the experience in higher education. Buckingham: SRHE/Open University Press. Rabardel, P., & Beguin, P. (2005). Instrument mediated activity: from subject development to anthropocentric design. Theoretical issues in ergonomic science, 6(5), 429-461. Reigeluth, C., & Carr-Chellman, A. (Eds.). (2009). Instructional design theories and models. Volume 3. New York: Routledge. Shuell, T. (1992) Designing instructional computing systems for meaningful learning in M. Jones & P. Winne (eds.) Adaptive Learning Environments, New York: Springer Verlag. Sorensen, E. (2009) The materiality of learning: technology and knowledge in educational practice, Cambridge: Cambridge University Press. Thagard, P. (2008) Hot thought: mechanisms and applications of emotional cognition, Cambridge MA: MIT Press. Turner, P. (2005) Affordance as context, Interacting with Computers, 17:787-800. Westberry, N., and Franken, M. (in press) Co-construction of knowledge in tertiary online settings: An ecology of resources perspective, Instructional Science. !

""!

AUTHOR BIOGRAPHIES ! Peter Goodyear: is Professor of Education, Australian Laureate Fellow and CoDirector of the CoCo Research Centre at the University of Sydney, Australia. He has been carrying out research in the field of learning and technology since the early 80s and has published seven books and almost 100 journal articles and book chapters. His most recent co-authored book was for Routledge (Students' experiences of e-learning in higher education: the ecology of sustainable innovation, with Rob Ellis, 2010). His research has taken place in the UK, mainland Europe and Australia and has been funded by the Australian Research Council, the UK Economic & Social Research Council, UK Government and Industry and the European Commission.

Lucila Carvalho: is a Postdoctoral Research Associate in the CoCo Research Centre at the University of Sydney, Australia. Her PhD research investigated the sociology of learning in/about design, and ways of practically implementing sociological principles into e-learning design. She has studied and carried out research in Australia, New Zealand, the UK and Brazil. She has presented her work at various international conferences in the fields of education, sociology, systemic functional linguistics, design and software engineering. Her most recent research has been published in Design Studies and she is co-editor (with Peter Goodyear) of the forthcoming book The architecture of productive learning networks. !

"#!

FIGURE ! !"#$%&'()'*'+,-%."-/0'1&.23%4'35'367&8.9'-1:'-8.";"."&9'

"$!